| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.21905805038335158, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004381161007667032, | |
| "grad_norm": 0.28300485014915466, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1534, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004381161007667032, | |
| "eval_loss": 0.178242489695549, | |
| "eval_runtime": 10.3376, | |
| "eval_samples_per_second": 9.383, | |
| "eval_steps_per_second": 4.74, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.008762322015334063, | |
| "grad_norm": 0.33447790145874023, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1934, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.013143483023001095, | |
| "grad_norm": 0.27208104729652405, | |
| "learning_rate": 6e-05, | |
| "loss": 0.1603, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.017524644030668127, | |
| "grad_norm": 0.34873825311660767, | |
| "learning_rate": 8e-05, | |
| "loss": 0.1852, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.02190580503833516, | |
| "grad_norm": 0.4304073750972748, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1692, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02190580503833516, | |
| "eval_loss": 0.12932729721069336, | |
| "eval_runtime": 10.3651, | |
| "eval_samples_per_second": 9.358, | |
| "eval_steps_per_second": 4.727, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02628696604600219, | |
| "grad_norm": 0.40466296672821045, | |
| "learning_rate": 0.00012, | |
| "loss": 0.114, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.03066812705366922, | |
| "grad_norm": 0.3059878349304199, | |
| "learning_rate": 0.00014, | |
| "loss": 0.1037, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.03504928806133625, | |
| "grad_norm": 0.20368129014968872, | |
| "learning_rate": 0.00016, | |
| "loss": 0.055, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03943044906900329, | |
| "grad_norm": 0.31183746457099915, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0435, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04381161007667032, | |
| "grad_norm": 0.2398935854434967, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0406, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04381161007667032, | |
| "eval_loss": 0.029431650415062904, | |
| "eval_runtime": 10.3988, | |
| "eval_samples_per_second": 9.328, | |
| "eval_steps_per_second": 4.712, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04819277108433735, | |
| "grad_norm": 0.28297099471092224, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 0.0256, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.05257393209200438, | |
| "grad_norm": 0.13293632864952087, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.0161, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.056955093099671415, | |
| "grad_norm": 0.238490030169487, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.0233, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.06133625410733844, | |
| "grad_norm": 0.1766541600227356, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0222, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.06571741511500548, | |
| "grad_norm": 0.08793684095144272, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.0064, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06571741511500548, | |
| "eval_loss": 0.018376410007476807, | |
| "eval_runtime": 10.3128, | |
| "eval_samples_per_second": 9.406, | |
| "eval_steps_per_second": 4.751, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0700985761226725, | |
| "grad_norm": 0.1723138689994812, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.0143, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.07447973713033954, | |
| "grad_norm": 0.11893691122531891, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.0115, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.07886089813800658, | |
| "grad_norm": 0.23403066396713257, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0339, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0832420591456736, | |
| "grad_norm": 0.4720830023288727, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.0228, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.08762322015334063, | |
| "grad_norm": 0.11689777672290802, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.0248, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08762322015334063, | |
| "eval_loss": 0.017750361934304237, | |
| "eval_runtime": 10.3124, | |
| "eval_samples_per_second": 9.406, | |
| "eval_steps_per_second": 4.752, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09200438116100766, | |
| "grad_norm": 0.1209622249007225, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.0109, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0963855421686747, | |
| "grad_norm": 0.04279917851090431, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.0042, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.10076670317634173, | |
| "grad_norm": 0.11789191514253616, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.0147, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.10514786418400876, | |
| "grad_norm": 0.28664666414260864, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.0328, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.10952902519167579, | |
| "grad_norm": 0.10409914702177048, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.0087, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.10952902519167579, | |
| "eval_loss": 0.0160684771835804, | |
| "eval_runtime": 10.3064, | |
| "eval_samples_per_second": 9.412, | |
| "eval_steps_per_second": 4.754, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.11391018619934283, | |
| "grad_norm": 0.08153662085533142, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0163, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.11829134720700986, | |
| "grad_norm": 0.12469758093357086, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.0166, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.12267250821467689, | |
| "grad_norm": 0.14660221338272095, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.0147, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.12705366922234393, | |
| "grad_norm": 0.13516534864902496, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.0156, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.13143483023001096, | |
| "grad_norm": 0.06466881185770035, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0121, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13143483023001096, | |
| "eval_loss": 0.015977900475263596, | |
| "eval_runtime": 10.3044, | |
| "eval_samples_per_second": 9.413, | |
| "eval_steps_per_second": 4.755, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13581599123767799, | |
| "grad_norm": 0.0801151692867279, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.0121, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.140197152245345, | |
| "grad_norm": 0.050880130380392075, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.0048, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.14457831325301204, | |
| "grad_norm": 0.12199495732784271, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.0158, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.14895947426067907, | |
| "grad_norm": 0.04026995226740837, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.008, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.1533406352683461, | |
| "grad_norm": 0.054878607392311096, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.0049, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.1533406352683461, | |
| "eval_loss": 0.015133237466216087, | |
| "eval_runtime": 10.3131, | |
| "eval_samples_per_second": 9.406, | |
| "eval_steps_per_second": 4.751, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.15772179627601315, | |
| "grad_norm": 0.09998418390750885, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.0126, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.16210295728368018, | |
| "grad_norm": 0.060268584638834, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.0054, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.1664841182913472, | |
| "grad_norm": 0.05094964802265167, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.0067, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.17086527929901424, | |
| "grad_norm": 0.07953735440969467, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.0135, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.17524644030668127, | |
| "grad_norm": 0.07781189680099487, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.0124, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.17524644030668127, | |
| "eval_loss": 0.01449812576174736, | |
| "eval_runtime": 10.3095, | |
| "eval_samples_per_second": 9.409, | |
| "eval_steps_per_second": 4.753, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1796276013143483, | |
| "grad_norm": 0.08530998229980469, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.0125, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.18400876232201532, | |
| "grad_norm": 0.07010316848754883, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0078, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.18838992332968238, | |
| "grad_norm": 0.06331722438335419, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.0105, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.1927710843373494, | |
| "grad_norm": 0.0915939211845398, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.0141, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.19715224534501644, | |
| "grad_norm": 0.05658958479762077, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.0076, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.19715224534501644, | |
| "eval_loss": 0.014185325242578983, | |
| "eval_runtime": 10.3071, | |
| "eval_samples_per_second": 9.411, | |
| "eval_steps_per_second": 4.754, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.20153340635268346, | |
| "grad_norm": 0.054891057312488556, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.0059, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.2059145673603505, | |
| "grad_norm": 0.04845288023352623, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.0045, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.21029572836801752, | |
| "grad_norm": 0.10850649327039719, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.0181, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.21467688937568455, | |
| "grad_norm": 0.060124583542346954, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.0067, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.21905805038335158, | |
| "grad_norm": 0.07994550466537476, | |
| "learning_rate": 0.0, | |
| "loss": 0.0124, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.21905805038335158, | |
| "eval_loss": 0.01418597437441349, | |
| "eval_runtime": 10.3302, | |
| "eval_samples_per_second": 9.39, | |
| "eval_steps_per_second": 4.743, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.69912849629184e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |